ligo- g030479-00-z lsc asis meeting 2003.03.20ligo scientific collaboration - university of...

LSC ASIS Meeting 2003.03.20 LIGO Scientific Collaboration - University of Wisconsin - Milwaukee

1

LIGO-G030479-00-Z

S2 Hardware Pulsar Injections

Bruce Allen, for the pulsar group, andPeter Shawhan, Szabi Marka, Scott Koranda


2

LIGO-G030479-00-Z

Why Do Hardware Injections?

• Provides end-to-end validation of search code:» Gain confidence about tricky things like floating point dynamic

range in filtering process» Helps algorithm/code developers in testing» Provide a fixed point of reference to return to

• Challenges in the pulsar case:» Realistic signals must last for hours (~107 cycles). This is unlike the

burst (seconds) and inspiral (tens of seconds) case.» Would like to avoid the “SB” scenario where the simulated signal

dominates the data» Getting the correct initial phase relationship at the different detector

sites can be tricky (specifics later)

• Simpler in pulsar case:» Calibration: signal is at a “single” frequency


3

LIGO-G030479-00-Z

S2 Pulsar Injection Parameters

• Signal is sum of two different pulsars, P1 and P2P1: Constant Intrinsic FrequencySky position: 0.3766960246 lattitude (radians)

5.1471621319 longitude (radians)Signal parameters are defined at SSB GPS time733967667.026112310 which corresponds to a wavefront passing:LHO at GPS time 733967713.000000000LLO at GPS time 733967713.007730720In the SSB the signal is defined byf = 1279.123456789012 Hzfdot = 0phi = 0A+ = 1.0 x 10-21

Ax = 0 [equivalent to iota=pi/2]

Note: At SSB GPS time 733967667.018380990 (wavefront passing LLO at GPS time 733967713.000000000) f is the same, because fdot=0, and phi=0.69546841529.

P2: Spinning DownSky position: 1.23456789012345 lattitude (radians)

2.345678901234567890 longitude (radians)Signal parameters are defined at SSB GPS time:SSB 733967751.522490380, which corresponds to awavefront passing:LHO at GPS time 733967713.000000000LLO at GPS time 733967713.001640320In the SSB at that moment the signal is defined byf=1288.901234567890123fdot = -10-8 [phase=2 pi (f dt+1/2 fdot dt^2+...)]phi = 0A+ = 1.0 x 10-21


Note: at SSB GPS time 733967751.520850180 a wavefront passes LLO at (slightly earlier) GPS time 733967713.000000000. At this time the signal is defined by f = 1288.901234567906525 (slightly higher)phi = 5.56655154939301191885 radiansfdot, A+ and Ax are unchanged


4

LIGO-G030479-00-Z

What Are These Two Different SSB GPS times?

P1: Constant Intrinsic FrequencySky position: 0.3766960246 lattitude (radians)

5.1471621319 longitude (radians)Signal parameters are defined at SSB GPS time733967667.026112310 which corresponds to a wavefront passing:LHO at GPS time 733967713.000000000LLO at GPS time 733967713.007730720In the SSB the signal is defined byf = 1279.123456789012 Hzfdot = 0phi = 0A+ = 1.0 x 10-21


Note: At SSB GPS time 733967667.018380990 (wavefront passing LLO at GPS time 733967713.000000000) f is the same, because fdot=0, and phi=0.69546841529.

#! /bin/bash

bc -l << EOFscale=20pi=4*a(1)f=1279.123456789012phi=2*pi*(.01838099-.02611231)*fPhi=phi+20*piprint "phi=", phi, "\n"EOF

phi=0.69546841529179172719

Note that in an April 14th email Bruce didn’t notice thatnsec had only 8 digits and incorrectly gave one of thesetimes as 733967667.26112310


5

LIGO-G030479-00-Z

How was simulated signal made?

• 12 hours of strain data was produced (with overall 1021 normalization factor) using LAL routines:

• 144 files (5 minutes each) were produced. Each file contains 16384*300+1 4-byte IEEE 754 floats:» Key 1234.5 (4 bytes)» Sample 0 xxxxxx (4 bytes)» Sample 1 yyyyyy (4 bytes)» …

• Time range was 00:00 –12:00 UTC April 10th• The 2.8 GB of injection data was shipped to each site

S2_pulsar_LHO_733968013.datS2_pulsar_LHO_733968313.datS2_pulsar_LHO_733968613.dat… and so on.


6

LIGO-G030479-00-Z

Simulated strain: Detector Response Function

April 10th 04:00 UTC is:LHO: 8 pm April 9th

LLO: 10 pm April 9th

April 10th 10:00 UTC is:LHO: 2 am April 10th

LLO: 4 am April 10th

simulated source ispassing near a zero ofthe antenna pattern.


7

LIGO-G030479-00-Z

How and When?

• More than 9 hours of pulsar injections into L1, H1 and H2» Start 18:19 PDT on April 9» Stop 04:03 PDT on April 10

• Instruments were in lock for almost the entire time• Pulsar plus calibration line summed into DARM_CTRL• ETM_X and ETM_Y used for other injections• Strain/DARM_CTRL calibrations worked out at 1284 Hz,

halfway between two signals. NOTE: THESE WERE SUBSEQUENTLY REVISED FOR H1 AND H2 – SO AT THOSE DETECTORS THE OVERALL STRAIN FACTOR WAS NOT 1.e-21!

• Injections started at 733974613 (01:50:00 UTC April 10) and continued until 734007889 (11:04:36 UTC April 10) with some minor interruptions (loss of lock, realignment, computer restarted because of lack of memory).


8

LIGO-G030479-00-Z

S2 Pulsar Hardware InjectionResults for P1 (Time Domain Code)

H1L1

These plots show the probability density functions obtained for h0, iota, phi and psi using the Markov Chain Monte Carlo method (N. Christensen, R. Dupuis, G. Woan).

They are all SHARPLY PEAKED at the correct values of h0, psi and iota. They are also sharply peaked at the wrong value of phi – but the same wrong value in H1, H2, L1. This is probably due to a typo (Bruce’s fault) in an email specifying the correct time origin for defining the phase.

The injected signal has been detected (modulo typo) at the correct values of the parameters!


9

LIGO-G030479-00-Z

Still to do…

• A quick look was done using the Frequency Domain method (F-statistic) within a few days after the run. The signals were seen at approximately the correct amplitude in all three instruments

• The Frequency Domain searches need to be repeated more carefully and better documented.

• The Time-Domain method still needs to look for the non-zero fdot signal. We also need to verify that with correct choice of initial GPS time, phi is peaked at zero (BA is almost 100% confident about this).


10

LIGO-G030479-00-Z

What about S3?• The pulgroup will try to provide a fast real-time

function that can be called, which will return h(t). Perhaps this can be used to do longer-term injections during S3.

• Uta Weiland (GEO Hannover) is writing a routine for this purpose for GEO pulsar injections.

• I’d like to know what the LIGO experimenters need, please.

• Discussion item: should we do long-term injections (weeks/months) during S3?

ligo- g030479-00-z lsc asis meeting 2003.03.20ligo scientific collaboration - university of...

Documents